Entanglement-Aware Anomaly Detection: Spin Density Matrices as Discovery Axes in ττ and tt̄

Zhang et al. (2025) show that with ML-based neutrino reconstruction one can measure the full τ+τ− spin density matrix and observe Bell nonlocality beyond 5σ. That’s a transformative measurement channel—so let’s turn it into an anomaly search axis. We propose to: (i) build robust estimators of spin density matrices and entanglement measures (concurrence, CHSH violations, CP-odd spin correlations) for ττ and tt̄ final states; (ii) integrate these into token-based generative models (Visive et al., 2025) that predict spin tokens/observables under the SM and flag deviations as anomalies; (iii) cross-relate anomalies with energy-correlator features (Lee et al., 2025) to separate spin-structure new physics from purely kinematic/QCD effects. This goes beyond kinematic outliers by explicitly targeting quantum information content that is highly sensitive to chiral, CP-violating, and contact-interaction effects. Novelty: whereas prior anomaly methods mostly operate on four-vectors or detector-level tokens, we use reconstructed quantum-state descriptors as first-class features, aided by ML for missing momentum inference. Impact: opens a new, complementary discovery axis with built-in interpretability—if an excess appears in entanglement or Bell metrics, we immediately learn about the chiral/CP nature of the underlying interaction.

References:

1. Event Tokenization and Next-Token Prediction for Anomaly Detection at the Large Hadron Collider. A. Visive, P. Moskvitina, C. Nellist, R. D. Austri, Sascha Caron Institute of Physics, U. Amsterdam, Amsterdam, The Netherlands, Nikhef, Dutch National Institute for Subatomic Physics, High Energy Physics, R. University, Nijmegen, Instituto de F'isica Corpuscular, IFIC-UVCSIC, Paterna, Spain. (2025).
2. Conformal collider physics meets LHC data. Kyle Lee, Bianka Meçaj, I. Moult (2022). Physical Review D.
3. Entanglement and Bell Nonlocality in $\tau^+ \tau^-$ at the LHC using Machine Learning for Neutrino Reconstruction. Yulei Zhang, Bai-Hong Zhou, Qibin Liu, T. Wu, Shu Li, Tao Han, Shih-Chieh Hsu, M. Low (2025).